From Middle Childhood to Adulthood Attachment

Abstract

Attachment theory argues that attachment styles developed during infancy and childhood will be relatively enduring and continue during adulthood. While research on adult attachment has been mounting in the past, less is known on the enduring effect of attachment styles with mother, father, and their peer or peers during childhood on adult romantic attachment styles. Past research in fact have obtained mixed findings on the stability of attachment styles from childhood to adulthood, especially in the context of Malaysian setting. In order to explore the stability of attachment styles from childhood to adulthood, a survey among 400 married individuals was conducted and partial least squares was employed for data analysis. The application of attachment theory in this study was established via two dimensions of anxiety and avoidance. The result indicated that respondents who exhibited high levels of anxiety and high levels of avoidance with mother, father, and their peers would display the same attachment styles with their romantic partners or spouses during adulthood. The findings provide conclusive evidence on the continuity and stability of attachment styles with stronger and enduring bonds observed for peer-adult attachment.

Keywords

childhood attachment styles adult romantic attachment styles anxiety avoidance stability

Originally, attachment is an abstract of the emotional bond between a caregiver and child. Subsequently, this concept has been extended to understand the nature of romantic relationships during adulthood. Many studies pointed out that the type of attachment children have toward their parents is strongly associated with the future quality of attachment in romantic relationships that the child would experience as an adult (Collins & Read, 1990; Hazan & Shaver, 1987). In other words, children are expected to adopt the same attachment style in which they were raised and eventually applying the same patterns of behavior in their romantic relationships (Baptist, Thompson, Norton, Hardy, & Link, 2012).

Although a wealth of evidence indicates the stability of attachment style over time, many recent scholars advocated that early attachment styles are not necessarily identical to what people experience in their later romantic relationships and might be changed according to situations (Crowell, Fraley, & Shaver, 1999; Waeger, 2007). Apparently, several attachment patterns might be the change caused by new experiences (Staik, 2013), due to the situations that the individual has to deal with (Hassan, 2008; Mikulincer & Shaver, 2007), exposure to new ideas (Crowell, Treboux, & Waters, 2002), marital interactions (Dinero, Conger, Shaver, Widaman, & Larsen-Rife, 2008), and spouse attachment (Davila, Karney, & Bradbury, 1999).

Adding more to the perplexity of the issue is the differential effect of attachment figures such as mother, father, and peer on adult attachment. Attachment theory suggested that as children started to build their relationship with parents, they also started to create more relationships with peers as they grow and develop bonds with other people outside the family (Fraley & Davis, 1997). While research on the attachment to father and mother has been widely documented, less is known on the effect of attachment to their peers on adult attachment.

Attachment orientation is known to have a profound effect on relational quality, conflict communication, and marital stability. Since it is a relatively stable trait developed from early infancy (Feeney & Collins, 2014), attachment style would endure until adulthood and exert significant impact on present-day romantic relationships both positively and negatively. This is an issue of concern in the context of Malaysian society, given the fact that the number of divorces in Malaysia has more than doubled in just 8 years from 2004 with the number of Malaysian couples getting divorced rose from 24,207 in 2004 to 56,760 in 2012, which is equivalent to a marriage breaking down every 10 min (Boo, 2014). Unlike non-Muslim marriage, whose data are systematically managed and kept by National Registration Department, little or outdated statistics are available for Muslim marriage due to the often use of manual recording to the data by the religious councils. The religious councils or better known as Syariah Judiciary Department Malaysia (Jabatan Kehakiman Syariah Malaysia) reported that the number of Muslim couples getting divorced rose by 2.3 times from 20,916 in 2004 to 47,740 in 2012. A slightly higher rate of divorce was observed for the non-Muslim couples. The Department of Statistics Malaysia via the National Registration Department indicated that the non-Muslim marriages broke down at a slightly higher rate from 3,291 divorces in 2004, increasing by 2.7 times to 9,020 cases in 2012. At the state level, statistics by the Department of Statistics recorded that the total number of divorce in the state of Penang in 2012 was 1,069 among Muslim and 1,019 among non-Muslim.

Although it is assumed to have universal significance, attachment theory in fact is deeply rooted in mainstream Western context and culture. The theory itself has “not been thoroughly tested in the crucible of human diversity” and a tendency to overlook on the impact of culture on attachment style and orientation (Rothbaum, Weisz, Pott, Miyake, & Morelli, 2000, p. 1102). Ng, Trusty, and Crawford (2005) further reiterated cross-cultural generalizability of these findings to non-Western contexts has yet to be systematically supported.

Considering the inconclusive empirical findings on the stability of attachment style from childhood to adulthood particularly in the context of Malaysian culture, this study sets out to explore the continuity of childhood and adulthood attachment and further aims to investigate the differential effect of childhood attachment with mother, father, and peer with romantic attachment during adulthood.

Attachment Theory

Attachment theory is a joint work of John Bowlby and Mary Ainsworth (Ainsworth & Bowlby, 1991; Bowlby, 1988). Drawing on concepts from ethology, cybernetics, information processing, developmental psychology, and psychoanalysis, Bowlby (1969) formulated the basic tenets of the theory. He revolutionized our thinking about a child’s ties to the mother and its disruption through separation, deprivation, and bereavement, which is known as the context of child–caregiver. Bowlby (1977) and other observers of both human and primate behavior noted that when an infant is healthy, alert, and unafraid in the presence of its mother, it seems interested in exploring and mastering the environment and in establishing contact with other family and community members. In this context, the mother serves as a secure base for a child. Research by Ainsworth, Blehar, Waters, and Wall (1978) suggests that a mother’s sensitivity and responsiveness to her infant’s signals and needs during the first year of life are important factors to determine child attachment.

On the other hand, Bowlby (1977) added that mothers who are slow or inconsistent in responding to their own infant’s cries or who regularly intrude on or interfere with their own infant’s desired activities (sometimes to force affection on the infant at a particular moment) produce infants who cry more than usual, explore less than usual (even in the mother’s presence), mingle attachment behavior with overt expressions of anger, and seem to appear generally anxious. If the mother consistently rebuffs or rejects their own infant’s attempts to establish physical contact, the infant may learn to avoid her. An infant’s response to the separation and reunion episode with the mother has formed three types of infant attachment: secure, anxious–ambivalent, and avoidant.

The relationship and interaction between child and caregiver builds a person’s cognitive framework, comprising mental representations for understanding the world, self, and others or also known as internal working model. The internal working model guides the child’s future, social, and emotional behavior (McLeod, 2007). According to Bowlby (1977), the child and caregiver relationships are the foundations that shape the internal representations of self and others. The early interaction between a child and caregivers can determine a child’s working models of “self” and “others” for their later attachment.

A vast body of literature has concentrated on mother–child attachment, while the outcomes associated with father–child attachment is understudied. It is seen that the mother is still a primary caregiver compared to the father who plays a secondary parenting role. Although Bowlby (1977) and Ainsworth et al. (1978) emphasized the importance of a secure attachment to one’s primary caregiver and a limited body of works on father’s attachment, there are some early attachment research which revealed that infants are likely to be distressed upon separation from the mother or father (e.g., Kotelchuck, 1976) and have direct attachment-related behavior toward both mothers and fathers upon reunion (Lamb, 1976). Infants seem to attach not just to those who actively care for them but to those they interact with regularly including their father. This can be explained by a study by Grossmann and Grossmann (2009), who showed that the sensitive supportiveness of mother and father and the acceptance of the child were the main predictors of a child’s internal working models in close relationships. Aside from the mother and father as prime caregivers, high-quality care such as attentive, responsive, affectionate, and stimulating from persons other than their parents (e.g., nanny) will lead to children flourishing cognitively (Belsky, 1999). The impact of nonmaternal caregivers (e.g., nanny) can be seen as larger toward children who live in less optimal family environment or difficult temperaments or genetic vulnerabilities (Vermeer & Bakermans-Kranenburg, 2008). Although originally attachment was thought to be most important between parents and their children, there is now evidence other caregivers can form attached relationships and have a big impact on a child’s life.

According to Ainsworth (1989), children’s relationship with parents continued to build and influence their internal working model during adolescence and adulthood. Individuals start to create more attachment relationships with peers as they grow older. Furman and Buhrmester (1992) pointed out that the relationship with peers grow more relevant as the children mature. During that time, they start to develop bonds with other people outside their family members. It is important to know that the relationship between children and peers is more equal in terms of giving and receiving compared to the relationship between children and their parents where parents generally give more than they receive (Hazan & Zeifman, 1994). The importance of the parents’ relationship with their children is obvious in the formation of the children’s relationship with their peers (Furman & Wehner, 1994; Gorrese & Ruggieri, 2013; Laible, Carlo, & Raffaelli, 2000; Meeus, Iedema, Maassen, & Engels, 2005).

Bowlby (1977) indicated that a person’s early attachment set the stage for adult attachment styles including with a spouse. The theory that was earlier concerned with parent–child attachment and peer attachment was later expanded to include adult romantic relationships. Bartholomew (1990) and Bartholomew and Horowitz (1991) proposed a different way of categorizing adult romantic attachment styles by introducing four types of attachment styles, namely, secure, dismissive, preoccupied, and fearful.

Due to the variants and extensions of questionnaire proposed on adult romantic attachment styles and which measures to be applied, Brennan, Clark, and Shaver (1998) made a successful attempt in creating a more sensitive and comprehensive adult attachment measure that incorporated more psychological constructs. From the model of self and others which identified four dimensions (secure, preoccupied, dismissive, and fearful), Brennan et al.’s (1998) findings suggested that individual differences in attachment can be most parsimoniously represented along two fundamental dimensions of anxiety and avoidance. On a similar note, several other researchers (Fraley & Spieker, 2003; Fraley & Waller, 1998; Mikulincer & Shaver, 2007) also suggested that attachment style is best conceptualized using dimensions of anxiety (those who are anxious and fear rejection) and avoidance (those uncomfortable with closeness).

By crossing these two dimensions of anxiety and avoidance, four attachment orientations of secure, preoccupied, dismissive, and fearful could be identified. According to Brennan et al.’s (1998) scale, an individual with the “secure” cluster scored low on both anxiety and avoidance. Those in the “preoccupied” cluster appeared high on anxiety and low on avoidance, while those in the “dismissing” cluster scored high on avoidance and low on anxiety. Those in the “fearful” cluster scored high on both anxiety and avoidance. To date, these two-dimensional measurements were adopted by other studies for the investigations of four attachment styles (Domingue & Mollen, 2009; Jurič, 2011; Lin, 2003; Russell, Baker, & McNulty, 2013; Steuber, 2005; Weger, 2006).

In the work of Yárnoz-Yaben (2010), they claimed that persons categorized as secure subjects have low anxiety and avoidance. Secure individuals tend to engage in direct communication and self-disclosure compared to insecure individuals because they are confident that their partners will respond supportively, be able to acknowledge distress, and are capable of handling negative feelings in a constructive way due to having experienced warmth and sensitive treatment by their caregivers (Johnson, 2003; Kobak & Sceery, 1988).

On the other hand, the preoccupied individuals are more likely to make demands, withdraw, or cling because they tend to believe that their partners will reject them or they will protest at the unresponsiveness from their partner (Johnson, 2003). With high level of anxiety but low level of avoidance, preoccupied subjects tend to demonstrate dependence and preoccupation in their affective relationships (Yárnoz-Yaben, 2010)

The dismissive individual with a high level of avoidance but low level of anxiety harbors a desire to remain independent and avoid committed relationships. A highly avoidant individual attains autonomy and a sense of self-worth at the expense of intimacy (Bartholomew, 1990). In the work of Yárnoz-Yaben (2010), they claimed that those with a highly avoidant style are most likely to avoid intimacy and deactivate the attachment system as a means of defense, with the aim of maintaining their invulnerable and self-sufficient image.

Those individuals categorized as fearful (high in anxiety and avoidance) limit their interpersonal relationships since they fear rejection and suffering that this may cause. According to Bartholomew (1990), both dismissive (avoidant) and fearful individuals were most likely to avoid intimacy and deactivate their attachment systems but for different reasons and in different ways. They might also be similar to preoccupied subjects particularly when they have negative views of both others and themselves (Guerrero, Andersen, & Afifi, 2001). According to Firestone (2013), fearful individuals see their relationship as they need to go toward others to get their needs met but at the same time fear being hurt.

Attachment Stability

Parental communication with children can have a big impact on children’s well-being and shape the quality of the children’s relationship with their parents and others. In other words, attachment is a predictor of a child’s later social and emotional outcomes. It is expected that attachment theory during childhood (e.g., the separation and subsequent reunion with caregivers) affects adults’ style of attachment (Apostolidou, 2006; Hazan & Shaver, 1987; Vorria et al., 2007).

According to Hazan and Shaver (1987), well-known scholars in attachment studies, romantic love is conceptualized as an attachment process that shows the continuity of infant–parent attachment. Ainsworth (1989) reiterated that the internal working model of infant attachment to parents would continue into adulthood even with a new attachment figure. Similarly, Firestone (2013) added that an attachment pattern established in early childhood would continue to function as a working model for relationships in adulthood. For example, if we grew up in an insecure attachment pattern during childhood, we may seek to duplicate a similar pattern, although we may dislike it.

According to Chen, Liu, and Li (2000) in most cultures, parental attitudes and behaviors toward the child might have a long-term impact on parent–child relationships and the child’s (mal)adaptive functioning. In other words, the adult romantic attachment is reflective of early attachment styles. Hence, six hypotheses have been proposed. First, anxiety with mother during childhood is a direct predictor of anxiety during romantic adulthood. Second, avoidance with mother during childhood is a direct predictor of avoidance during romantic adulthood. Third, anxiety with father during childhood is a direct predictor of anxiety during romantic adulthood. The fourth hypothesis is avoidance with father during childhood is a direct predictor of avoidance during romantic adulthood.

As identified by Furman and Wehner (1994), children’ view of friendships are consistent with their views of parent–child relationships. It seems reasonable to assume that social skills learned by children from their parents tend to generalize to their peer relations (Ma & Huebner, 2008). Indeed, the relationships with peers may influence their later romantic attachment styles. The behavior of peers and the way peers interact with them are factors which contribute to later attachment styles (Kerns, Klepac, & Cole, 1996; Lieberman, Doyle, & Markiewicz, 1999). Based on our review of the literature, we formulated and tested Hypotheses 5 and 6. Anxiety with a peer during childhood is a direct predictor of anxiety during romantic adulthood, and avoidance with a peer during childhood is a direct predictor of avoidance during romantic adulthood.

Method

Sample

Technique-wise, a quantitative survey is applied in this study in order to achieve its objective. Individual respondents were selected mainly from the state of Penang, Malaysia, using a purposive sampling technique. Penang was chosen as it suitably represents Malaysian population characteristics. Being one of the most industrialized and urbanized state in Malaysia, Penang population characteristics reflect rich mixtures of different ethnic groups with various demographic and socioeconomic backgrounds that mirror the actual Malaysia population. Penang statistic indicated that the total number of population of Penang in 2017 is 1.75 million (https://penanginstitute.org). Of this, 648,999 were married with Chinese made up the highest number with 321,585 and this is followed by Malay at 258,492, Indian at 66,679, and other ethnic groups at 2,243. Penang also has recorded the highest population density in Malaysia—6,696 people per square kilometer due to high urbanization and fast economic growth compared to other states in the country. Due to its vibrant economic growth, Penang enjoys modern transportation services, equal access to education, and advanced health facilities.

Based on Krejcie and Morgan (1970), the sample size guidelines for the population size of over 75,000–1,000,000 is 384 (N = 648, 999, S = 384). Thus, the final sample was rounded up to 400. From 400 individuals (S = 400) of the sampling size, the researcher divided the respondents equally according to their ethnicity such as Chinese (195), Malay (155), and Indian (50). The respondents then were divided equally according to gender such as Chinese (male = 100, female = 95), Malay (male = 76, female = 79), and Indian (male = 25, female = 25). There was no age and duration of marriage limit criteria in the selection of respondents. However, efforts were made to ensure that the respondents belonged to one of the four family lifecycles (preparental, parental, launching, and postparental), which consists of Chinese (male = 25 × 4, female = 24 × 4, total = 195), Malay (male = 19 × 4, female = 20 × 4, total = 155), and Indian (male = 6 × 4, female = 6 × 4, total = 50). In this study, married individuals were chosen as the respondents because they are perceived to be in a stable relationship.

The percentages of male and female respondents are 50.3% and 49.8%, respectively, representing an equal gender proportion. In terms of race or ethnic group, the respondents are mainly among Chinese at 48.5% followed by Malays at 38.8% and Indian at 12.8%.

Measures

The questionnaire was developed from Experiences in Close Relationships Inventory (ECR) by Brennan et al. (1998). ECR is one of the most commonly used self-report instruments of romantic attachment. However, the ECR is also designed to assess attachment patterns in a variety of close relationships, not only in romantic relationships. Measures of attachment consist of two variables, namely, anxiety and avoidance. These variables were measured based on 36 items (18 items for each variable) in order to determine four categories of adult attachment based on model of self and model of others. The questionnaire was distributed using two versions of English and Malay languages. The questionnaire was translated from English to Malay, the official language of Malaysia, and in turn translated to English to ensure validity and reliability. The questionnaire in both English and Malay versions was further assessed for its validity by an expert panel. An expert panel, comprising two lecturers from Department of Communication, was asked to provide feedback. Upon receipt of feedback, the translated questions were revised. To be specific, the respondents were required to answer the questionnaire based on their current relationship with their spouse and recall their past relationship with their mothers, fathers, and peers during their middle childhood period. According to Collins (1984) and Brown (2015), middle childhood is between 6 and 12 years of age. A growing area of attachment research is the work investigating the role of attachment in middle childhood (Brumariu & Kerns, 2008; McCormick, O’Connor, & Barnes, 2016). The present review focuses exclusively on attachment relationships in middle childhood. At this stage, emergent cognitive abilities enable children to handle more complex intellectual problem-solving and to better understand reciprocal social relationships than they could in early childhood (Blume, 2014).

Accuracy of retrospective memory is a key challenge in any research involving recall or recollection of past memory. Feeney and Cassidy (2003) and Dykas, Woodhouse, Ehrlich, and Cassidy (2010) have asserted on general attachment consistency over time by reiterating that a person reconstructed their memory (e.g., conflict with parents) by their internal working models of attachment. For example, if a person’s internal working model is secure, he should be inclined to reconstruct his memory with other persons as positive because he is relying on secure internalized attachment representations and vice versa. This is consistent with studies by Pasupathi (2001) and Tversky and Marsh (2000), which stated that people recall events from memory in ways that reflect their views and perceptions. However, Scharfe and Bartholomew (1998, p. 221) asserted, otherwise, that the consistency of attachment representation over time would vary based on the attachment style itself, with “secure individuals more likely to present a coherent story of the past and current relationship,” while insecure attachment is often associated with an inaccurate recall. Kirkpatrick and Hazan’s (1994) study examined accuracy of retrospective reports of attachment categories over a period of 4 years, and results reported that secure individuals were more likely to maintain their attachment category than were insecure individuals.

In order to reduce the effect of bias and inaccuracy in retrospective memory of early attachment style, Scharfe and Bartholomew (1998) recommended the use of continuous attachment rating of Likert-type scale of 1–7 instead of categorical-based attachment classification which requires the respondents to choose the most relevant attachment styles (secure, fearful, dismissive, or preoccupied) that corresponds to their own attachment description. The present study has adopted the continuous rating of attachment dimensions following this recommendation.

After conducting a pilot test, 7 items of anxiety scale were removed due to low reliability value. The remaining scale comprised 11 items. From 18 items in avoidance scale, 6 items were deleted due to the same reason, leaving 12 remaining items for avoidance. All of the remaining items were found to be reliable after the removal of weak items. According to Field (2009), the most vital part in Cronbach’s α analysis is determining how each item individually contributes to the reliability of the questionnaire. Item deletion should be considered to increase the consistency and stability of the measure. As recommended by Raubenheimer (2004, p. 61), sequential removal of items must be conducted to maximize reliability and validity with “the single item which would lead to the highest increase in reliability is to be removed at each particular step.” Each scale also would “require at least 4 items to be properly identified” (Raubenheimer, 2004, p. 60). In the present study, items were removed in sequential order, and the items which remained after this process were found to contribute the highest reliability value.

Data Analysis

This study was tested using partial least squares (PLS). The PLS is relevant to this study based on the characteristics mentioned by Chin and Newsted (1999) where PLS is prediction-oriented and is a causal modeling approach aimed at maximizing the explained variance of the dependent latent constructs. In order to examine the model in this study, the process involved two stages. The first stage is the assessment of the measurement model and followed by the assessment and evaluation of the structural model.

Result

Measurement Model

The analysis of measurement model is to ensure the reliability and validity of the indicators for each construct before conducting further analysis and deriving conclusion of the structural model. This study comprises eight constructs. Six constructs represent attachment styles during childhood: anxiety with mother, avoidance with mother, anxiety with father, avoidance with father, anxiety with peer, and avoidance with peers. Two constructs represent adult romantic attachment: anxiety with spouse and avoidance with spouse. Each construct has multiple indicators or measurement items. After the analysis of measurement model, the number of indicators for each construct has been significantly reduced. Anxiety with mother has two indicators, avoidance with mother has four, anxiety with father has two, avoidance with father has four, anxiety with peer has four, avoidance with peer has six, anxiety with spouse has three, and avoidance with spouse has six. Urbach and Ahlemann (2010) compiled the validation guidelines set by other researchers for testing the internal consistency reliability, indicator reliability, convergent validity, and discriminant validity by applying standard decision rules of reflective measurement model. Internal consistency reliability is used to test the reliability of the construct, while convergent validity and discriminant validity to assess the construct validity.

To measure internal consistency, the reliability of measures analysis through composite reliability and Cronbach’s α. Results in Table 1 demonstrated satisfactory item reliability, with values larger than .6 judged as acceptable (Bagozzi & Yi, 1988). Second, the smart PLS bootstrap resampling procedure (e.g., 500 resamples) was assigned to measure convergent validity, which had created loading, composite reliability, and average variance extracted (AVE). The composite reliability was well above the commonly used .6 cutoff of all eight constructs (Bagozzi & Yi, 1988). The result in Table 1 also showed the AVE was in the range above the recommended value of .50 (Fornell & Larcker, 1981; Hair, Black, Babin, Anderson, & Tatham, 2006).

Table 1.

Loadings, Internal Consistency Reliabilities, and AVE From Measures.

Latent Variables	Question Items	Cronbach’s α	Composite Reliability	Main Loading	AVE
Anxiety mother	I often want to merge completely with my mother, and this sometimes scares her away.	.694	.867	.875	.765
Anxiety mother	I find that my mother doesn’t want to get as close as I would like.	.694	.867	.875
Avoidance mother	I get uncomfortable when my mother wants to be very close to me.	.820	.881	.782	.649
	I want to get close to my mother, but I keep pulling back.			.811
	I am nervous when my mother gets too close to me.			.812
	I prefer not to be too close to my mother.			.817
Anxiety father	I often want to merge completely with my father, and this sometimes scares him away.	.837	.924	.921	.859
Anxiety father	My desire to be very close to my father sometimes scares him away.	.837	.924	.933
Avoidance father	Just when my father starts to get close to me, I find myself pulling away.	.829	.886	.808	.661
	I am nervous when my father gets too close to me.			.801
	I try to avoid getting too close to my father.			.814
	I prefer not to be too close to my father.			.828
Anxiety peer	I often want to merge completely with them, and this sometimes scares them away.	.804	.872	.828	.632
	My desire to be very close to my peer sometimes scares him/her away.			.849
	Sometimes I feel that I force my peer to show more feeling and commitment.			.687
	I find that my peer doesn’t want to get as close as I would like.			.805
Avoidance peer	Just when my peer starts to get close to me, I find myself pulling away.	.830	.874	.772	.538
	I get uncomfortable when my peer wants to be very close to me.			.741
	I want to get close to my peer, but I keep pulling back			.756
	I am nervous when my peer gets too close to me.			.749
	I try to avoid getting too close to my peer.			.715
	I prefer not to be too close to my peer.			.772
Anxiety spouse	I often want to merge completely with my spouse, and this sometimes scares him/her away.	.831	.899	.864	.747
	My desire to be very close to my spouse sometimes scares him/her away.			.907
	I find that my spouse doesn’t want to get as close as I would like.			.819
Avoidance spouse	Just when my spouse starts to get close to me, I find myself pulling away.	.853	.891	.757	.577
	I get uncomfortable when my spouse wants to be very close to me.			.771
	I want to get close to my spouse, but I keep pulling back.			.749
	I am nervous when my spouse gets too close to me.			.759
	I try to avoid getting too close to my spouse.			.744
	I prefer not to be too close to my spouse.			.757

Note. N = 400. AVE = average variance extracted.

To show the discriminant validity, items correlated most strongly with their intended construct (as shown by loadings and cross loadings). According to Fornell and Larcker (1981), the items should load more strongly on their own construct in the model, while the average variance shared between each construct and its measures should be larger than the squared correlations among the latent variable and other variable. The result in Table 2 indicated all square roots of AVE exceeded the off-diagonal values in their corresponding row and column, indicating acceptable discriminant validity.

Table 2.

Intercorrelations of Constructs.

	AN-F	AN-M	AN-P	AN-S	AV-F	AV-M	AV-P	AV-S
AN-F	.927
AN-M	.854	.875
AN-P	.716	.722	.795
AN-S	.796	.772	.774	.864
AV-F	.689	.725	.618	.595	.813
AV-M	.686	.780	.628	.633	.790	.806
AV-P	.562	.620	.700	.570	.679	.703	.733
AV-S	.620	.692	.632	.712	.711	.741	.685	.759

Note. N = 400. AN-F = anxiety father; AN-M = anxiety mother; AN-P = anxiety peer; AN-S = anxiety spouse; AV-F = avoidance father; AV-M = avoidance mother; AV-P = avoidance peer; AV-S = avoidance spouse.

Structural Model

The result of validity in the structural model can be evaluated to test the research hypotheses. The process of checking the structural model’s validity involves the evaluation of path coefficient (β) and coefficient of determination (R²). Hypothesis 1 was supported: Anxiety with mother during childhood is a direct predictor of anxiety during romantic adulthood (β = .196, t = 2.766, p < .01). Hypothesis 2 was supported: Avoidance with mother during childhood is a direct predictor of avoidance during romantic adulthood (β = .269, t = 3.626, p < .001). Hypothesis 3 was supported: Anxiety with father during childhood is a direct predictor of anxiety during romantic adulthood (β = .379, t = 6.187, p < .001). Hypothesis 4 was supported: Avoidance with father during childhood is a direct predictor of avoidance during romantic adulthood (β = .198, t = 2.762, p < .01). Hypothesis 5 was supported: Anxiety with peer during childhood is a direct predictor of anxiety during romantic adulthood (β = .392, t = 7.214, p < .001). Finally, Hypothesis 6 also was supported: Avoidance with peer during childhood is a direct predictor of avoidance during romantic adulthood (β = .207, t = 3.056, p < .01; see Table 3). The model in Figure 1 explained the high proportion of variance between anxiety and avoidance with mother, father, and peer and anxiety with spouse (R² = .730). Additionally, it also showed good results between anxiety and avoidance with mother, father, and peer and avoidance with spouse (R² = .639).

Table 3.

Partial Least Square Structural Equation Model Results: Direct Effect of Childhood Attachment (Mother, Father, and Peer) and Romantic Adulthood Attachment (Spouse).

Hypotheses	Relationship	Path Coefficient (β)	Standard Error	t Value	Result
H1	Anxiety mother → anxiety spouse	.196	.071	2.766**	Supported
H2	Avoidance mother → avoidance spouse	.269	.074	3.626***	Supported
H3	Anxiety father → anxiety spouse	.379	.061	6.187***	Supported
H4	Avoidance father → avoidance spouse	.198	.072	2.762**	Supported
H5	Anxiety peer → anxiety spouse	.392	.054	7.214***	Supported
H6	Avoidance peer → avoidance spouse	.207	.068	3.056**	Supported

*p < .05. **p < .01. ***p < .001.

Figure 1.

The Coefficient of determination (R² values). AN-F = anxiety father; AN-M = anxiety mother; AN-P = anxiety peer; AN-S = anxiety spouse; AV-F = avoidance father; AV-M = avoidance mother; AV-P = avoidance peer; AV-S = avoidance spouse.

Discussion

This study sets out to explore the continuity and stability of childhood and adulthood attachment and further aims to investigate the differential effect of childhood attachment with mother, father, and peer with romantic attachment during adulthood in a Malaysian context, specifically in the state of Penang. In general, the results showed that the respondents’ early attachment during childhood with mother, father, and their peers remains the same and continues into their romantic adulthood with their spouses.

Attachment literature emphasizes that the relationship with both parents is a predictive to the subsequent adult attachment style including with one’s spouse (Ainsworth, 1989; Collins & Read, 1990; Firestone, 2013; Fraley, 2010; Hazan & Shaver, 1987). Additionally, Bowlby (1977) and Ainsworth and Bell (1990) suggested that the development of attachment theory is formulated from the idea of the relationship between infants and parents and based on the strong emotional bond that parents always have with their child/children. The stability of attachment style from childhood to adulthood was strongly evident in a longitudinal study by Waters, Merrick, Treboux, and Albersheim (2000). In their study, the result indicates that 64% of 50 individuals admitted that their attachment classifications were stable in 20 years.

Comparatively, anxious individuals who described their relationship with their parents during childhood as rejecting, cold, unfair, and inconsistent would tend to replicate the same attachment style during adulthood, while those who described their relationship with their parents during childhood as loving, inviting, warm, fair, and consistent will enjoy similar attachment style with their spouse (Brennan & Shaver, 1990; Vorria et al., 2007).

To be more specific, children who have parents who treated them in an inconsistent or intrusive way will become anxious and fearful of not knowing what to expect. Later at relationships during adulthood, they may be unsure, hesitant, and worried about their spouse due to the nature of relationship that they had with their parents who may be available one moment and rejecting the next (Bretherton & Munholland, 2008; Segal & Jaffe, 2015). On the other hand, a highly avoidant adult may have a negative stigma toward their intimate relationship and tend to have difficulties in recalling important events and would basically describe their early history of rejection with their parents compared to those who enjoy secure attachment with their spouses who described their relationship with parents as mutually loving and giving (Apostolidou, 2006; Main & Solomon, 1990; Segal & Jaffe, 2015).

The current research has indicated that attachment with peer is the most significant attachment as compared to the mother and father. This is because the researchers believed that an adolescent’s relationship with peers is the starting process of transition toward becoming an adult. The study has shown that peers are the first persons who an adolescent derive emotional support from outside the house (Paterson, Field, & Pryor, 1994) and having a more equal level of giving and receiving relationship compared to relationship with parents who give more than they receive (Hazan & Zeifman, 1994). This study could state that the relationship with peers is the process of relationship development in teaching an adolescent toward maturity, while the tendency to maintain the same attachment until adulthood is strong.

The present study has addressed the issue of stability of attachment styles from childhood to adulthood, by providing strong empirical evidence on the enduring effect of attachment styles among married individuals in Penang. Results also indicated greater attachment stability with peers thereby influencing the respondents’ attachment orientation during adulthood with their spouses.

Limitation, Future Research, and Conclusion

Although the findings from this study indicate interesting and noteworthy results, the study has encountered a number of limitations. The first limitation is in terms of its sample. The present study is conducted among respondents in Penang, Malaysia, which may impact the generalizability of the findings. Future research should be expanded to cover other areas in Malaysia. Random sampling technique should be applied, so that the data can be generalized and well-distributed. Another shortcoming in this study is related to the issue of accuracy of retrospective memory from childhood to adulthood. According to Baldwin (1995) and Fiske and Taylor (1991), memories may not always be accurate particularly on a person’s interpersonal experience. Over time, memory recall may degrade or exaggerate or subject to bias to current experience. Although a person is able to recall their past experiences, factors such as mood, needs, and types of personal experience (such as distress vs. nondistress experience) may affect the accuracy of these retrospective memories. For example, Dykas et al. (2010) found that secure adolescents reconstructed conflict with their parents more favorably over time, while insecure adolescents displayed less favorable recollection of past conflict with their parents. Simpson, Rholes, and Winterheld (2010) similarly found that a highly avoidant adult reported to experience less parental support during childhood distress. Future research should consider using a panel study to collect information on the same individuals at different points in time to observe on the enduring effect of childhood attachment to adulthood attachment.

The limitations notwithstanding, the study notably contributes to a limited number of studies on attachment stability in a non-Western context. The results of the present study are particularly significant because they shed new light on the enduring effect of childhood attachment with mother, father, and peer on adult romantic attachment in the current sociocultural context of Malaysia. This particular research finding points to the need for a major relook into the approach used by the National Population and Family Development Board in its family and marital counseling sessions. The attachment theory should be used as the main framework in understanding the pivotal links between child–caregiver and their subsequent impact on adults’ relational, emotional, and social well-being. Findings from this study can be included in the materials presented during caregiver training courses or prenatal education classes in order to improve child–caregiver interaction, to foster attachment security, and to learn when and how to respond to children’s sign of distress. The same material must also be embedded in the premarital counseling program before the issuance of a marriage certificate/license and must be incorporated as the main tool in couple therapy and clinical practice during postmarital counseling session conducted by the marital counselors and therapists in Malaysia.

Footnotes

Authors’ Note

This article is based on the doctoral dissertation.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Ainsworth

M. D. S.

(1989). Attachments beyond infancy. American Psychologist, 44, 709–716.

Ainsworth

M. D. S.

Bell

S. M.

(1990). Attachment, exploration and separation: Illustrated by the behavior of one-year-olds in a strange situation. Child Development, 41, 49–67.

Ainsworth

M. D. S.

Blehar

Waters

Wall

(1978). Patterns of attachment. Hillsdale, NJ: Lawrence Erlbaum.

Ainsworth

M. D. S.

Bowlby

(1991). An ethological approach to personality development. American Psychologist, 46, 331–341.

Apostolidou

(2006). Are childhood experiences with parents linked to feelings in romantic relationships during adulthood? The New School Psychology Bulletin Volume, 4, 63–85.

Bagozzi

R. P.

(1988). On the evaluation of structural equation models. Journal of the Academy of Marketing Science, 16, 74–94.

Baldwin

M. W.

(1995). Relational schemas and cognition in close relationships. Journal of Social and Personal Relationships, 12, 547–552.

Baptist

J. A.

Thompson

D. E.

Norton

A. M.

Hardy

N. R.

Link

C. D.

(2012). The effects of the intergenerational transmission of family emotional processes on conflict styles: The moderating role of attachment. American Journal of Family Therapy, 40, 56–73.

Bartholomew

(1990). Avoidance of intimacy: An attachment perspective. Journal of Social and Personal Relationships, 7, 147–178.

10.

Bartholomew

Horowitz

L. M.

(1991). Attachment styles among young adults: A test of a four-category model. Journal of Personality and Social Psychology, 61, 226–244.

11.

Belsky

(1999). Modern evolutionary theory and patterns of attachment. In Cassidy

Shaver

P. R.

(Eds.), Handbook of attachment: Theory, research and clinical applications. New York, NY: Guilford.

12.

Blume

L. B.

(2014). Middle childhood. Retrieved June 18, 2014, from http://www.oxfordbibliographies.com/view/document/obo-9780199791231/obo-9780199791231-0142.xml

13.

Boo

S. L.

(2014). One divorce in Malaysia every 10 minutes. Retrieved May 6, 2018, from www.malaymail.com/s/628027/one-divorce-in-malaysia-every-10-minutes

14.

Bowlby

(1969). Attachment and loss: Vol. I. Attachment. New York, NY: Basic Books.

15.

Bowlby

(1977). The making and breaking of affectional bonds. London, England: Tavistock.

16.

Bowlby

(1988). Developmental psychiatry comes of age. American Journal of Psychiatry, 145, 1–10.

17.

Brennan

Clark

Shaver

(1998). Self-report measurement of adult attachment. In Simpson

J. A.

Rholes

W. S.

(Eds.), Attachment theory and close relationships (pp. 46–76). New York, NY: Guilford Press.

18.

Brennan

K. A.

Shaver

P. R.

(1990). Dimensions of adult attachment, affect regulation, and romantic relationship functioning. Personality and Social Psychology Bulletin, 21, 267–283.

19.

Bretherton

Munholland

K. A.

(2008). Internal working models in attachment relationships: elaborating a central construct. In Cassidy

Shaver

(Eds.), Handbook of attachment: Theory, research and clinical applications (2nd ed., pp. 102–127). London, England: Guilford Press.

20.

Brown

(2015). Middle and late childhood (6-12 Years). Retrieved June 15, 2017, from https://prezi.com/bvj0ia3rmv0y/middle-and-late-childhood-6-12-years/

21.

Brumariu

L. E.

Kerns

K. A.

(2008). Mother-child attachment and social anxiety symptoms in middle childhood. Journal of Applied Development Psychology, 29, 393–402.

22.

Chen

Liu

(2000). Parental warmth, control, and indulgence and their relations to adjustment in Chinese children: A longitudinal study. Journal of Family Psychology, 14, 401–419.

23.

Chin

W. W.

Newsted

P. R.

(1999). Structural equation modelling analysis with small samples using partial least squares. In Hoyle

R. H.

(Ed.), Statistical strategies for small sample research (pp. 307–341). Thousand Oaks, CA: Sage.

24.

Collins

W. A.

(1984). Development during middle childhood: The years from six to twelve. Washington, DC: National Academy Press.

25.

Collins

N. L.

Read

S. J.

(1990). Adult attachment, working models, and relationship quality in dating couples. Journal of Personality and Social Psychology, 58, 644–663.

26.

Crowell

J. A.

Fraley

R. C.

Shaver

P. R.

(1999). Measurement of individual differences in adolescent and adult attachment. In Cassidy

Shaver

P. R.

(Eds.), Handbook of attachment: Theory, research, and clinical applications. New York, NY: Guilford Press.

27.

Crowell

J. A.

Treboux

Waters

(2002). Stability of attachment representations: The transition to marriage. Developmental Psychology, 38, 467–479.

28.

Davila

Karney

B. R.

Bradbury

T. N.

(1999). Attachment change processes in the early years of marriage. Journal of Personality and Social Psychology, 76, 783–802.

29.

Dinero

R. E.

Conger

R. D.

Shaver

P. R.

Widaman

K. F.

Larsen-Rife

(2008). Influence of family of origin and adult romantic partners on romantic attachment security. Journal of Family Psychology, 22, 622–632. doi:10.1037/a0012506

30.

Domingue

Mollen

(2009). Attachment and conflict communication in adult romantic relationships. Journal of Social and Personal Relationships, 26, 678–696. doi:10.1177/0265407509347932

31.

Dykas

M. J.

Woodhouse

S. S.

Ehrlich

K. B.

Cassidy

(2010). Do adolescents and parents reconstruct memories about their conflict as a function of adolescent attachment? Child Development, 81, 1445–1459. doi:10.1111/j.1467-8624.2010.01484.x

32.

Feeney

B. C.

Cassidy

(2003). Reconstructive memory related to adolescent–parent conflict interactions: The influence of attachment-related representations on immediate perceptions and changes in perceptions over time. Journal of Personality and Social Psychology, 85, 945–955.

33.

Feeney

B. C.

Collins

N. L.

(2014). Couple relationships. Retrieved April 2, 2015, from http://family.jrank.org/pages/118/Attachment.html

34.

Field

(2009). Discovering statistics using SPSS. London, England: Sage.

35.

Firestone

(2013). How your attachment style impacts your relationship. Retrieved July 26, 2016, from http://www.psychalive.org/how-your-attachment-style-impacts-your-relationship/

36.

Fiske

S. T.

Taylor

S. E.

(1991). Social Cognition. New York, NY: McGraw-Hill.

37.

Fornell

Larcker

D. F.

(1981). Evaluating structural equation models with unobservable variables and measurement errors. Journal of Marketing Research, 18, 39–50.

38.

Fraley

R. C.

(2010). A brief overview of adult attachment theory and research. Retrieved September 19, 2016, from https://internal.psychology.illinois.edu/∼rcfraley/attachment.htm

39.

Fraley

R. C.

Davis

K. E.

(1997). Attachment formation and transfer in young adults’ close friendships and romantic relationships. Personal Relationships, 4, 131–144.

40.

Fraley

R. C.

Spieker

S. J.

(2003). Are infant attachment patterns continuously or categorically distributed? A taxometric analysis of strange situation behavior. Developmental Psychology, 39, 387–404.

41.

Fraley

R. C.

Waller

N. G.

(1998). “Adult attachment patterns: A test of the typological model. In Simpson

J. A.

Rholes

W. S.

(Eds.), Attachment theory and close relationships (pp. 77–114). New York, NY: Guilford.

42.

Furman

Buhrmester

(1992). Age and sex differences in perceptions of networks of personal relationships. Child Development, 63, 103–115.

43.

Furman

Wehner

E. H.

(1994). Romantic views: Toward a theory of adolescent romantic relationships. In Montemayor

Adams

G. R.

Gullotta

T. P

(Eds.), Advances in adolescent development: An annual book series (Vol. 6, pp. 168–195). Thousand Oaks, CA: Sage.

44.

Gorrese

Ruggieri

(2013). Peer attachment and self-esteem: A meta-analytic review. Personality and Individual Differences, 55, 559–568.

45.

Grossmann

K. E.

(2009). The impact of attachment to mother and father and sensitive support of exploration at an early age on children’s psychosocial development through young adulthood. Encyclopedia on Early Childhood Development. Regensburg, Germany: University of Regensburg.

46.

Guerrero

L. K.

Andersen

P. A.

Afifi

W. A.

(2001). Close encounters: communicating in relationship. New York, NY: McGraw-Hill.

47.

Hair

J. F.

Black

W. C.

Babin

B. J.

Anderson

R. E.

Tatham

L. R.

(2006). Multivariant data analysis. Upper Saddle River, NJ: Pearson International Edition.

48.

Hassan

S. A.

(2008). Maternal quality time conception: A phenomenological study of employed mothers. Serdang, Malaysia: Universiti Putra Malaysia Press.

49.

Hazan

Shaver

(1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52, 511–524.

50.

Hazan

Zeifman

(1994). Sex and the psychological tether. In Bartholomew

Perlman

(Eds.), Advances in personal relationships: Attachment processes in adulthood (Vol. 5, pp. 151–177). London, UK: Jessica Kingsley.

51.

Johnson

S. M.

(2003). Attachment theory: A guide for couple therapy. In Johnson

S. M.

Whiffen

V. E.

(Eds.), Attachment processes in couple and family therapy (pp. 103–123). New York, NY: Guilford Press.

52.

Jurič

(2011). Difference in relationship satisfaction and adult attachment in married and cohabitating couples. Anthropos, 1–2, 51–69.

53.

Kerns

K. A.

Klepac

Cole

A. K.

(1996). Peer relationships and preadolescents’ perceptions of security in the mother–child relationship. Developmental Psychology, 32, 457–466.

54.

Kirkpatrick

L. A.

Hazan

(1994). Attachment styles and close relationships: A four-year prospective study. Personal Relationships, 1, 123–142.

55.

Kobak

R. R.

Sceery

(1988). Attachment in late adolescence: Working models, affect regulations, and representations of self and others. Child Development, 59, 135–146.

56.

Kotelchuck

(1976). The infant’s relationship to the father: Experimental evidence. In Lamb

M. E.

(Ed.), The role of the father in child development (pp. 329–344). New York, NY: Wiley.

57.

Krejcie

R. V.

Morgan

D. W.

(1970). Determining sample size for research activities. Educational and Psychological Measurement, 30, 607–610.

58.

Laible

D. J.

Carlo

Raffaelli

(2000). The differential relations of parent and peer attachment to adolescent adjustment. Journal of Youth and Adolescence, 29, 45–59.

59.

Lamb

M. E.

(1976). Interactions between two-year-olds and their mothers and fathers. Psychological Reports, 38, 447–450.

60.

Lieberman

Doyle

Markiewicz

(1999). Developmental patterns of security of attachment to mother and father in late childhood and early adolescence: Associations with peer relations. Child Development, 70, 202–213.

61.

Lin

(2003). The association between adult attachment styles and conflict resolution in romantic relationship. The American Journal of Family Therapy, 31, 143–157. doi:10.1080/01926180390167142

62.

C. Q.

Huebner

E. S.

(2008). Attachment relationship and adolescents’ life satisfaction: Some relationships matter more to girls than boys. Psychology in the Schools, 45, 177–190.

63.

Main

Solomon

(1990). Procedures for identifying infants as disorganized/disoriented during Ainsworth strange situation. In Greenberg

M. T.

Cicchetti

Cummings

E. M.

(Eds.), Attachment in preschool years: Theory, research and intervention (pp. 121–160). Chicago, IL: University of Chicago Press.

64.

McCormick

M. P.

O’Connor

E. E.

Barnes

S. P.

(2016). Mother–child attachment styles and math and reading skills in middle childhood: The mediating role of children’s exploration and engagement. Early Childhood Research Quarterly, 36, 295–306.

65.

McLeod

(2007). Bowlby’s attachment theory. Retrieved April 24, 2018, from https://www.simplypsychology.org/bowlby.html

66.

Meeus

Iedema

Maassen

Engels

(2005). Separation-individuation revisited: On the interplay of parent–adolescent relations, identity and emotional adjustment in adolescence. Journal of Adolescence, 28, 89–106.

67.

Mikulincer

Shaver

P. R.

(2007). Attachment in adulthood: Structure, dynamics, and change. New York, NY: Guilford Press.

68.

K. M.

Trusty

Crawford

(2005). A cross-cultural validation of the attachment style questionnaire: A Malaysian pilot study. The Family Journal: Counseling and Therapy for Couples and Families, 13, 416–426. doi:10.1177/1066480705278471

69.

Paterson

J. E.

Field

Pryor

(1994). Adolescents’ perceptions of their attachment relationships with their mothers, fathers, and friends. Journal of Youth and Adolescence, 23, 579–600.

70.

Pasupathi

(2001). The social construction of the personal past and its implications for adult development. Psychological Bulletin, 127, 651–672.

71.

Penang Statistics. (2017). Retrieved August 1, 2018, from https://penanginstitute.org/resources/data-centre/97-quarterly-penang-statistics

72.

Rothbaum

Weisz

Pott

Miyake

Morelli

(2000). Attachment and culture security in the United States and Japan. American Psychologist, 55, 1093–1104. doi:10.1037//0003-066X.55.10.1093

73.

Raubenheimer

(2004). An item selection procedure to maximise scale reliability and validity. Journal of Industrial Psychology, 30, 59–64.

74.

Russell

V. M.

Baker

L. R.

McNulty

J. K.

(2013). Attachment insecurity and infidelity in marriage: Do studies of dating relationships really inform us about marriage? Journal of Family Psychology, 27, 242–251.

75.

Scharfe

Bartholomew

(1998). Do you remember? Recollections of adult attachment patterns. Personal Relationships, 5, 219–234.

76.

Segal

Jaffe

(2015). Attachment and adult relationships. Retrieved March 3, 2015, from http://www.helpguide.org/articles/relationships/attachment-and-adult-relationships.htm

77.

Simpson

J. A.

Rholes

W. S.

Winterheld

H. A.

(2010). Attachment working models twist memories of relationship events. Psychological Science, 21, 252–259. doi:10.1177/0956797609357175

78.

Staik

(2013). Relating styles and love relationships. Retrieved March 31, 2013, from http://blogs.psychcentral.com/relationships/2013/08/the-impact-of-childhood-attachment-styles-on-couple-relationships-2-of-2/

79.

Steuber

K. R.

(2005). Adult attachment, conflict style and relationship satisfaction: A comprehensive model. Master’s dissertation, University of Delaware. Retrieved from http://library.allanschore.com/docs/kelisteuberthesis.pdf

80.

Tversky

Marsh

E. J.

(2000). Biased retellings of events yield biased memories. Cognitive Psychology, 40, 1–38.

81.

Urbach

Ahlemann

(2010). Structural equation modeling in information systems research using partial least square. Journal of Information Technology Theory and Application, 11, 5–40.

82.

Vermeer

H. J.

Bakermans-Kranenburg

M. J.

(2008) Attachment to mother and nonmaternal care: Bridging the gap. Attachment & Human Development, 10, 263–273.

83.

Vorria

Vairami

Gialaouzidas

Kotroni

Koutra

Markou

… Pantoleon

(2007). Romantic relationships, attachment styles, and experiences of childhood. Hellenic Journal of Psychology, 4, 281–309.

84.

Waeger

(2007). A qualitative study of how attachment and separation affects the life course of individuals. Retrieved March 5, 2014, from http://david-waeger.blogspot.com/2007/10/qualitative-study-of-how-attachment-and.html

85.

Waters

Merrick

S. K.

Treboux

Albersheim

(2000). Attachment security in infancy and early adulthood: A 20 –year longitudinal study. Child Development, 71, 684–689.

86.

Weger

Jr (2006). Associations among romantic attachment, argumentativeness, and verbal aggressiveness in romantic relationships. Argumentation and Advocacy, 43, 29–40.

87.

Yárnoz-Yaben

(2010). Attachment style and adjustment to divorce. The Spanish Journal of Psychology, 13, 210–219.